snapsy wrote:
Based on DxOMark's data, the IV's ISO 12,800 is equal to the D3s's ISO 25,600. That means the X will have a one-stop improvement over the D3s if Chuck's metrics haven't changed and if we're talking about raw performance rather than Digic-processed performance (ie noise reduction).
1 stop at the pixel level or 2 stops if down ressed to 12MP.
omarlyn wrote:
I find it curious that Canon did not incorporate an articulated LCD screen with the 1DX. Sport shooters & journalist in particular could use such a screen for shooting from the over-head position and it's also good in videography to flip/tilt the screen as per your shooting needs.
Omar
I'll bet it doesn't fit in with their build requirements or sealing requirements. It could be a weak point and not liable to take the hammering these are built for.
uz2work wrote:
According to Chuck Westfall, who should know more about Canon products than any Canon employee in the US, the AF cut off for the 1D X is, in fact, f5.6.
Below, I've pasted Westfall's reply to Arthur Morris, who was asking about the AF cut off.
"AF is unavailable on the EOS-1D X if the maximum aperture reported to the camera through the electronic lens mount is smaller than f/5.6. This is a lower specification than previous EOS-1 series DSLRs. On the plus side, consider the fact that with most f/4 lenses including the 400 DO, 500/4L IS and IS II, and 600/4L IS and IS II, you now have 41 cross-type AF points plus color and face detection, whereas you had no cross-type points and no color or face detection during AF with previous EOS-1 series DSLRs using the same lenses, not to mention a significantly wider AF coverage area from left to right."
I've seen similar confirmations on at least 3 websites.
Shit! That sucks big time. Given this is pixel challenged, a 1.4x will often be needed. Guess I won't be using my 100-400L. Also means no AF with f/4 lenses and 2x TC.
omarlyn wrote:
I find it curious that Canon did not incorporate an articulated LCD screen with the 1DX. Sport shooters & journalist in particular could use such a screen for shooting from the over-head position and it's also good in videography to flip/tilt the screen as per your shooting needs.
Omar
Pixel Perfect wrote:
I'll bet it doesn't fit in with their build requirements or sealing requirements. It could be a weak point and not liable to take the hammering these are built for.
Ah, excellent point…besides the weather sealing, I can just imagine a sports photographer & photo journalist running down a field or hallway with cameras & lenses all jostling about and one of those screens (that got left angled out) just snaps right off!
PetKal wrote:
* Same lens/TC combo , same geometry of light impingement:
-1DMkIV allows center point AF at f/8.
-1DX with more sensitive AF points doesn't allow AF at f/8.
Using any additional AF points is less relevant to this. I could have also suggested that they might have lowered the single point AF to f/11 in 1DX.
That sure sounds like an oddity to me.
* Canon probably knows very well that TC use which pushes nominal aperture to f/8 is relatively common, not sporadic at all.
It's not at all an oddity. When they're talking about aperture sensitivity they're talking about geometry sensitivity, not light sensitivity. So the design choices they made in making the sensor still cross type at f4 meant they're no longer at some threshold to be certified as of sufficient sensitivity at f/8. That said, they also claim to have improved the light sensitivity.
The second part is much more the issue, and my gut instinct says that it's certainly non zero though it won't be that common as a factor in a body purchase, given that the only situation this becomes a hard limit is when the combo in question is an 800/5.6 +TC or 600/4 + 2x TC (in the set of all 1D users likely to seriously consider this as a new body purchase). If you are one of those, it is a major issue - I just don't think there are enough people for whom it is an issue AND for whom this would tip a purchase decision from a yes to a no balanced against the other advantages the 1DX brings.
Of course, I could be horribly wrong.
Edit:
I don't see a reason why Canon could not have gone with f/8 centre and had the surrounding points as they are currently, except for the semantic incongruity of having the centre point not be the best with wide aperture lenses.
reno.peterson wrote:
Does this camera have Microfocus Adjust feature...
well...
AF Microadjustment now gives the option to apply two different adjustment values for zooms, one for each end of the zoom range, and you can also apply different adjustment values to two units of the same lens model, differentiating between the two based on their serial number.
PetKal wrote:
* Same lens/TC combo , same geometry of light impingement:
-1DMkIV allows center point AF at f/8.
-1DX with more sensitive AF points doesn't allow AF at f/8.
Using any additional AF points is less relevant to this. I could have also suggested that they might have lowered the single point AF to f/11 in 1DX.
That sure sounds like an oddity to me.
* Canon probably knows very well that TC use which pushes nominal aperture to f/8 is relatively common, not sporadic at all.
That really sounds strange yes, given that the X is the new flagship model, and the other possible flavors of high end cameras are not likely to get a better AF system.
Either there was a technical difficulty we don't understand, or we will be forced to use both FF and APS-C for birds an such in the future.
ausemmao wrote:
Working in lower light levels is not quite the same thing as working with narrower aperture lenses. Lower light levels mess with the quantity of light the sensor has to work with. Narrower aperture size messes with the geometry of the incoming light as well as the light level.
When you have a smaller aperture the light rays are now effectively paraxial, so coming in almost parallel to the optical axis. I wouldn't call that "messing" with the geometry of the light. With f/2.8 you can get rays coming in at large angles to the optical axis. Many of these rays can be vignetted, can suffer more distortion etc than paraxial rays.
Pixel Perfect wrote:
When you have a smaller aperture the light rays are now effectively paraxial, so coming in almost parallel to the optical axis. I wouldn't call that "messing" with the geometry of the light. With f/2.8 you can get rays coming in at large angles to the optical axis. Many of these rays can be vignetted, can suffer more distortion etc than paraxial rays.
That's the point. For an AF sensor (unless I have grossly misunderstood the technique involved) you do not want the incident light to be parallel to the lens axis, as that hampers the ability to differentiate phase when the light hits the sensors. The vignetting and distortion don't have an insurmountable effect in the centre of the imaging area (indeed, chromatic aberration is one of the hurdles they mention overcoming in improving AF performance) though they do at the edges which is likely why there aren't AF sensors there as yet.
If 1DX were to support AF with the smallest nominal aperture of f/5.6 only, that would turn out to be a problem. Yes, similar to what many of us have been experiencing for years with non 1D camera bodies.
I know of a bunch of people who have been using the following combos habitually: 500+2xTC, 600+2xTC and 800+ 1.4xTC.
(The only one that I might use on occasion is 800 + 1.4xTC MkIII because IQ is still good and the combo is plenty long.)
ausemmao wrote:
That's the point. For an AF sensor (unless I have grossly misunderstood the technique involved) you do not want the incident light to be parallel to the lens axis, as that hampers the ability to differentiate phase when the light hits the sensors. The vignetting and distortion don't have an insurmountable effect in the centre of the imaging area (indeed, chromatic aberration is one of the hurdles they mention overcoming in improving AF performance) though they do at the edges which is likely why there aren't AF sensors there as yet.
Ok, that makes sense bout the phase. But I knew this is why you can't have extreme AF point locations near corners or edges of sensor.
wickerprints wrote:
"Microadjustment with zoom lenses has also been made easier. In the past it was only possible to register one microadjustment setting for each lens. However, with the EOS-1D X it's now possible to make adjustments for both the wideangle and telephoto settings of a zoom lens. This will help to ensure that whatever focal length you use with a zoom, your images are accurately focused."
Does it allow you to select the specific focal length to adjust, or simply the long and short end?
If I were a 1DIV owner I think I'd be a little POed. The 1DX seems aimed at replacing it, not so much the 1DsIII. Maybe they'll make the 1DsIII owners drool over the one to be announced next. From what little I know right now I wouldn't be the least bit tempted to replace my 1Ds MarkIII with a 1D X.
Looks like this camera is born out of 7D. The AF modules(AF zone selection etc: except 61 AF points) and DIGIC 4 for metering and AF control (Looks like more expanded version 7D AF system) and they use Dual DIGIC 5+ image processors. And now no AF at F/8 and Flash Syc speed 1/250...I am keeping my mark IV.
Tony Rogers wrote:
ISO runs from ISO 100 to ISO 51200 in "normal" + two levels of expansion: 102400 (limit of the 1D Mark IV) and 204800 (new and shiny!). That's 10 stops in normal and 12 stops including the expansions.
1Dx expands down to ISO 50 so thats 13 stops including expansion.
campyone wrote:
If I were a 1DIV owner I think I'd be a little POed. The 1DX seems aimed at replacing it, not so much the 1DsIII. Maybe they'll make the 1DsIII owners drool over the one to be announced next. From what little I know right now I wouldn't be the least bit tempted to replace my 1Ds MarkIII with a 1D X.
Why? I own a 1DIV and I am looking forward to this camera, the best of both worlds for me.
Do you guys really think that Canon is going to release a Flagship model that will not AF with apertures smaller than f/8. Come on, that is ridiculous. Or am I simply misunderstanding the issue.
well...
